Implementation of image compression using Fractal Image Compression and neural networks for MRI images

Fractal Image compression (FIC) is a new compression technique in the spatial domain. It is based on block based image compression technique which, detects and codes the existing similarities between different regions in the image. The main drawback of FIC is that the encoding time is comparatively very high, whereas the decompression is very fast. The encoding process consists of the searching for an appropriate domain block for each range block. In order to decrease the search space and time for encoding, it is proposed to device an FIC algorithm for MRI images using back propagation neural network algorithm. During the encoding process, an MRI image is divided into range and domain blocks. Then the index of each range block is given as input to the trained system and this result in a set of best matched domain blocks. An exhaustive search is done to extract the best matched domain block for all range blocks and this results in a set of range block index and its best matched domain block index. The neural network is then trained with these resultant values. This trained net is now used to compress other MRI images which lead to a very less encoding time. During the decoding phase, the transformation parameters are recursively applied to any arbitrary initial image, which then converges to the fractal image after some iteration. The simulation results show that the performance of this Neural Network based FIC is greatly.